Evolution of openEuler Embedded ROS Robot Application Ecosystem and Expansion of Deep Learning Scenarios
[Project Description]:
openEuler Embedded is mainly targeted at embedded scenarios, currently supporting armNN. The goal of this project is to expand the ROS application ecosystem and help the evolution of embedded ROS versions and breakthroughs in deep learning scenarios.
[Project difficulty]:
Advanced
[Output Standards]:
Adapt and transplant a ROS demo that requires AI (related to image recognition, any demo can be chosen, such as AI deep learning patrol, see material for details) based on originbot, and demonstrate it on a real machine (hardware should be prepared, at least a Raspberry Pi 4B+ and any UVC camera, originbot may not be needed, meaning that no requirements for controlling the inference result after broadcasting the result through ROS node and being observed correctly from the PC side). It's better to use armNN as the inference framework, and provide materials.
All ROS2 software package versions need to use the humble version and related software packages need to be adapted (the ROS of openEuler Embedded is currently using the foxy version, which is about to stop maintenance by the upstream community, which will involve upgrading the version of existing ROS core software packages that have not switched, and transplanting and introducing new software packages for deep learning applications).
[Technical Requirements]:
Familiar with Yocto framework, and have a certain understanding of embedded Linux systems and cross-compilation.
Familiar with deep learning framework applications.
Evolution of openEuler Embedded ROS Robot Application Ecosystem and Expansion of Deep Learning Scenarios
[Project Description]:
openEuler Embedded is mainly targeted at embedded scenarios, currently supporting armNN. The goal of this project is to expand the ROS application ecosystem and help the evolution of embedded ROS versions and breakthroughs in deep learning scenarios.
[Project difficulty]:
Advanced
[Output Standards]:
Adapt and transplant a ROS demo that requires AI (related to image recognition, any demo can be chosen, such as AI deep learning patrol, see material for details) based on originbot, and demonstrate it on a real machine (hardware should be prepared, at least a Raspberry Pi 4B+ and any UVC camera, originbot may not be needed, meaning that no requirements for controlling the inference result after broadcasting the result through ROS node and being observed correctly from the PC side). It's better to use armNN as the inference framework, and provide materials.
All ROS2 software package versions need to use the humble version and related software packages need to be adapted (the ROS of openEuler Embedded is currently using the foxy version, which is about to stop maintenance by the upstream community, which will involve upgrading the version of existing ROS core software packages that have not switched, and transplanting and introducing new software packages for deep learning applications).
[Technical Requirements]:
Familiar with Yocto framework, and have a certain understanding of embedded Linux systems and cross-compilation.
Familiar with deep learning framework applications.
Evolution of openEuler Embedded ROS Robot Application Ecosystem and Expansion of Deep Learning Scenarios
[Project Description]:
openEuler Embedded is mainly targeted at embedded scenarios, currently supporting armNN. The goal of this project is to expand the ROS application ecosystem and help the evolution of embedded ROS versions and breakthroughs in deep learning scenarios.
[Project difficulty]:
Advanced
[Output Standards]:
Adapt and transplant a ROS demo that requires AI (related to image recognition, any demo can be chosen, such as AI deep learning patrol, see material for details) based on originbot, and demonstrate it on a real machine (hardware should be prepared, at least a Raspberry Pi 4B+ and any UVC camera, originbot may not be needed, meaning that no requirements for controlling the inference result after broadcasting the result through ROS node and being observed correctly from the PC side). It's better to use armNN as the inference framework, and provide materials.
All ROS2 software package versions need to use the humble version and related software packages need to be adapted (the ROS of openEuler Embedded is currently using the foxy version, which is about to stop maintenance by the upstream community, which will involve upgrading the version of existing ROS core software packages that have not switched, and transplanting and introducing new software packages for deep learning applications).
[Technical Requirements]:
Familiar with Yocto framework, and have a certain understanding of embedded Linux systems and cross-compilation.
Familiar with deep learning framework applications.
Evolution of openEuler Embedded ROS Robot Application Ecosystem and Expansion of Deep Learning Scenarios
[Project Description]:
openEuler Embedded is mainly targeted at embedded scenarios, currently supporting armNN. The goal of this project is to expand the ROS application ecosystem and help the evolution of embedded ROS versions and breakthroughs in deep learning scenarios.
[Project difficulty]:
Advanced
[Output Standards]:
Adapt and transplant a ROS demo that requires AI (related to image recognition, any demo can be chosen, such as AI deep learning patrol, see material for details) based on originbot, and demonstrate it on a real machine (hardware should be prepared, at least a Raspberry Pi 4B+ and any UVC camera, originbot may not be needed, meaning that no requirements for controlling the inference result after broadcasting the result through ROS node and being observed correctly from the PC side). It's better to use armNN as the inference framework, and provide materials.
All ROS2 software package versions need to use the humble version and related software packages need to be adapted (the ROS of openEuler Embedded is currently using the foxy version, which is about to stop maintenance by the upstream community, which will involve upgrading the version of existing ROS core software packages that have not switched, and transplanting and introducing new software packages for deep learning applications).
[Technical Requirements]:
Familiar with Yocto framework, and have a certain understanding of embedded Linux systems and cross-compilation.
Familiar with deep learning framework applications.
Evolution of openEuler Embedded ROS Robot Application Ecosystem and Expansion of Deep Learning Scenarios
[Project Description]:
openEuler Embedded is mainly targeted at embedded scenarios, currently supporting armNN. The goal of this project is to expand the ROS application ecosystem and help the evolution of embedded ROS versions and breakthroughs in deep learning scenarios.
[Project difficulty]:
Advanced
[Output Standards]:
Adapt and transplant a ROS demo that requires AI (related to image recognition, any demo can be chosen, such as AI deep learning patrol, see material for details) based on originbot, and demonstrate it on a real machine (hardware should be prepared, at least a Raspberry Pi 4B+ and any UVC camera, originbot may not be needed, meaning that no requirements for controlling the inference result after broadcasting the result through ROS node and being observed correctly from the PC side). It's better to use armNN as the inference framework, and provide materials.
All ROS2 software package versions need to use the humble version and related software packages need to be adapted (the ROS of openEuler Embedded is currently using the foxy version, which is about to stop maintenance by the upstream community, which will involve upgrading the version of existing ROS core software packages that have not switched, and transplanting and introducing new software packages for deep learning applications).
[Technical Requirements]:
Familiar with Yocto framework, and have a certain understanding of embedded Linux systems and cross-compilation.
Familiar with deep learning framework applications.
Evolution of openEuler Embedded ROS Robot Application Ecosystem and Expansion of Deep Learning Scenarios
[Project Description]:
openEuler Embedded is mainly targeted at embedded scenarios, currently supporting armNN. The goal of this project is to expand the ROS application ecosystem and help the evolution of embedded ROS versions and breakthroughs in deep learning scenarios.
[Project difficulty]:
Advanced
[Output Standards]:
Adapt and transplant a ROS demo that requires AI (related to image recognition, any demo can be chosen, such as AI deep learning patrol, see material for details) based on originbot, and demonstrate it on a real machine (hardware should be prepared, at least a Raspberry Pi 4B+ and any UVC camera, originbot may not be needed, meaning that no requirements for controlling the inference result after broadcasting the result through ROS node and being observed correctly from the PC side). It's better to use armNN as the inference framework, and provide materials.
All ROS2 software package versions need to use the humble version and related software packages need to be adapted (the ROS of openEuler Embedded is currently using the foxy version, which is about to stop maintenance by the upstream community, which will involve upgrading the version of existing ROS core software packages that have not switched, and transplanting and introducing new software packages for deep learning applications).
[Technical Requirements]:
Familiar with Yocto framework, and have a certain understanding of embedded Linux systems and cross-compilation.
Familiar with deep learning framework applications.
Evolution of openEuler Embedded ROS Robot Application Ecosystem and Expansion of Deep Learning Scenarios
[Project Description]:
openEuler Embedded is mainly targeted at embedded scenarios, currently supporting armNN. The goal of this project is to expand the ROS application ecosystem and help the evolution of embedded ROS versions and breakthroughs in deep learning scenarios.
[Project difficulty]:
Advanced
[Output Standards]:
Adapt and transplant a ROS demo that requires AI (related to image recognition, any demo can be chosen, such as AI deep learning patrol, see material for details) based on originbot, and demonstrate it on a real machine (hardware should be prepared, at least a Raspberry Pi 4B+ and any UVC camera, originbot may not be needed, meaning that no requirements for controlling the inference result after broadcasting the result through ROS node and being observed correctly from the PC side). It's better to use armNN as the inference framework, and provide materials.
All ROS2 software package versions need to use the humble version and related software packages need to be adapted (the ROS of openEuler Embedded is currently using the foxy version, which is about to stop maintenance by the upstream community, which will involve upgrading the version of existing ROS core software packages that have not switched, and transplanting and introducing new software packages for deep learning applications).
[Technical Requirements]:
Familiar with Yocto framework, and have a certain understanding of embedded Linux systems and cross-compilation.
Familiar with deep learning framework applications.
Evolution of openEuler Embedded ROS Robot Application Ecosystem and Expansion of Deep Learning Scenarios
[Project Description]:
openEuler Embedded is mainly targeted at embedded scenarios, currently supporting armNN. The goal of this project is to expand the ROS application ecosystem and help the evolution of embedded ROS versions and breakthroughs in deep learning scenarios.
[Project difficulty]:
Advanced
[Output Standards]:
Adapt and transplant a ROS demo that requires AI (related to image recognition, any demo can be chosen, such as AI deep learning patrol, see material for details) based on originbot, and demonstrate it on a real machine (hardware should be prepared, at least a Raspberry Pi 4B+ and any UVC camera, originbot may not be needed, meaning that no requirements for controlling the inference result after broadcasting the result through ROS node and being observed correctly from the PC side). It's better to use armNN as the inference framework, and provide materials.
All ROS2 software package versions need to use the humble version and related software packages need to be adapted (the ROS of openEuler Embedded is currently using the foxy version, which is about to stop maintenance by the upstream community, which will involve upgrading the version of existing ROS core software packages that have not switched, and transplanting and introducing new software packages for deep learning applications).
[Technical Requirements]:
Familiar with Yocto framework, and have a certain understanding of embedded Linux systems and cross-compilation.
Familiar with deep learning framework applications.
Evolution of openEuler Embedded ROS Robot Application Ecosystem and Expansion of Deep Learning Scenarios
[Project Description]:
openEuler Embedded is mainly targeted at embedded scenarios, currently supporting armNN. The goal of this project is to expand the ROS application ecosystem and help the evolution of embedded ROS versions and breakthroughs in deep learning scenarios.
[Project difficulty]:
Advanced
[Output Standards]:
Adapt and transplant a ROS demo that requires AI (related to image recognition, any demo can be chosen, such as AI deep learning patrol, see material for details) based on originbot, and demonstrate it on a real machine (hardware should be prepared, at least a Raspberry Pi 4B+ and any UVC camera, originbot may not be needed, meaning that no requirements for controlling the inference result after broadcasting the result through ROS node and being observed correctly from the PC side). It's better to use armNN as the inference framework, and provide materials.
All ROS2 software package versions need to use the humble version and related software packages need to be adapted (the ROS of openEuler Embedded is currently using the foxy version, which is about to stop maintenance by the upstream community, which will involve upgrading the version of existing ROS core software packages that have not switched, and transplanting and introducing new software packages for deep learning applications).
[Technical Requirements]:
Familiar with Yocto framework, and have a certain understanding of embedded Linux systems and cross-compilation.
Familiar with deep learning framework applications.
Evolution of openEuler Embedded ROS Robot Application Ecosystem and Expansion of Deep Learning Scenarios
[Project Description]:
openEulerEmbeddedis mainly targeted atembedded scenarios, currentlysupporting armNN. The goal of this project is to expand the ROS applicationecosystem and help the evolution of embedded ROS versions and breakthroughs indeep learning scenarios.
[Project difficulty]:
Advanced
[Output Standards]:
Adapt and transplant a ROS demo that requires AI (related to image recognition, any demo can be chosen, such as AI deep learning patrol, see material for details) based on originbot, and demonstrate it on a real machine (hardware should be prepared, at least a Raspberry Pi 4B+ and any UVC camera, originbot may not be needed, meaning that no requirements for controlling the inference result after broadcasting the result through ROS node and being observed correctly from the PC side). It's better to use armNN as the inference framework, and provide materials.
All ROS2 software package versions need to use the humble version and related software packages need to be adapted (the ROS of openEuler Embedded is currently using the foxy version, which is about to stop maintenance by the upstream community, which will involve upgrading the version of existing ROS core software packages that have not switched, and transplanting and introducing new software packages for deep learning applications).
[Technical Requirements]:
Familiar with Yocto framework, and have a certain understanding of embedded Linux systems and cross-compilation.
Familiar with deep learning framework applications.
openEuler Embedded is mainly targeted at embedded scenarios, currently supporting armNN. The goal of this project is to expand the ROS application ecosystem and help the evolution of embedded ROS versions and breakthroughs in deep learning scenarios.
【Difficulty】：
Advanced
【Output Requirements】：
1、Adapt and transplant a ROS demo that requires AI (related to image recognition, any demo can be chosen, such as AI deep learning patrol, see material for details) based on originbot, and demonstrate it on a real machine (hardware should be prepared, at least a Raspberry Pi 4B+ and any UVC camera, originbot may not be needed, meaning that no requirements for controlling the inference result after broadcasting the result through ROS node and being observed correctly from the PC side). It's better to use armNN as the inference framework, and provide materials.
2、All ROS2softwarepackage versions needto use the humble version andrelated software packages need to be adapted (the ROS of openEuler Embeddedis currently using the foxy version, which is about to stop maintenance by theupstream community, which will involve upgrading the version of existing ROS core software packages that have not switched, and transplanting and introducing new software packages for deep learning applications).
【Technical Requirements】：
1、Familiar with Yocto framework, and have a certain understanding of embedded Linux systems and cross-compilation.
2、Familiar with deep learning framework applications.
openEuler Embedded is mainly targeted at embedded scenarios, currently supporting armNN. The goal of this project is to expand the ROS application ecosystem and help the evolution of embedded ROS versions and breakthroughs in deep learning scenarios.
【Difficulty】：
Advanced
【Output Requirements】：
1、Adapt and transplant a ROSdemo that requires AI (related to image recognition, any demo can bechosen,such asAIdeep learningpatrol, see material for details) based on originbot, and demonstrate it on areal machine (hardware should be prepared, at least a Raspberry Pi 4B+ and any UVC camera, originbot may not be needed, meaning that no requirements for controlling the inference result after broadcasting the result through ROS node and being observed correctly from the PC side). It's better to use armNN as the inference framework, and provide materials.
2、All ROS2 software package versions need to use the humble version and related software packages need to be adapted (the ROS of openEuler Embedded is currently using the foxy version, which is about to stop maintenance by the upstream community, which will involve upgrading the version of existing ROS core software packages that have not switched, and transplanting and introducing new software packages for deep learning applications).
【Technical Requirements】：
1、Familiarwith Yocto framework, andhavea certain understanding of embedded Linux systems and cross-compilation.
2、Familiar with deep learning framework applications.
Evolution of openEuler Embedded ROS Robot Application Ecosystem and Expansion of Deep Learning Scenarios
【Difficulty】：
Advanced
【Description】：
openEuler Embedded is mainlytargeted at embedded scenarios, currently supporting armNN. The goalof thisprojectisto expand theROS application ecosystem and help the evolution of embedded ROS versions andbreakthroughs in deep learning scenarios.
【Output Requirements】：
1、Adapt and transplant a ROS demo that requires AI (related to image recognition, any demo can be chosen, such as AI deep learning patrol, see material for details) based on originbot, and demonstrate it on a real machine (hardware should be prepared, at least a Raspberry Pi 4B+ and any UVC camera, originbot may not be needed, meaning that no requirements for controlling the inference result after broadcasting the result through ROS node and being observed correctly from the PC side). It's better to use armNN as the inference framework, and provide materials.
2、All ROS2software package versionsneedto use the humble version and related software packages need to be adapted (the ROS of openEuler Embedded is currently using the foxy version, which is about to stop maintenance by the upstream community, which will involve upgrading the version of existing ROS core software packages that have not switched, and transplanting and introducing new software packages for deep learning applications).
【TechnicalRequirements】：
1、Familiar with Yocto framework, and have a certain understanding of embedded Linux systems and cross-compilation.
2、Familiar with deep learning framework applications.
